Get Started

YOLO Experiment Tracking and Visualization with Weights & Biases

Object detection models like Ultralytics YOLO26 have become integral to many computer vision applications. However, training, evaluating, and deploying these complex models introduce several challenges. Tracking key training metrics, comparing model variants, analyzing model behavior, and detecting issues require significant instrumentation and experiment management.



Watch: How to use Ultralytics YOLO26 with Weights and Biases

This guide showcases Ultralytics YOLO26 integration with Weights & Biases for experiment tracking, best-model artifacts, and visualization of model performance. It also includes instructions for setting up the integration, training, and visualizing results using Weights & Biases' interactive features.

Already have YOLO checkpoints logged in W&B? Import their projects into Ultralytics Platform from Settings > Integrations > Weights & Biases.

Weights & Biases#

Weights & Biases is an MLOps platform designed for tracking, visualizing, and managing machine learning experiments. It features automatic logging of training metrics for full experiment reproducibility, an interactive UI for streamlined data analysis, and efficient model management tools for deploying across various environments.

YOLO26 Training with Weights & Biases#

You can use Weights & Biases to bring efficiency and automation to your YOLO26 training process. The integration allows you to track experiments, compare models, and make data-driven decisions to improve your computer vision projects.

Installation#

To install the required packages, run:

Installation
# Install the required packages for Ultralytics YOLO and Weights & Biases
pip install -U ultralytics wandb

# Enable W&B logging for Ultralytics
yolo settings wandb=True

For detailed instructions and best practices related to the installation process, be sure to check our YOLO26 Installation guide. While installing the required packages for YOLO26, if you encounter any difficulties, consult our Common Issues guide for solutions and tips.

Configuring Weights & Biases#

After installing the necessary packages, the next step is to set up your Weights & Biases environment. This includes creating a Weights & Biases account and obtaining the necessary API key for a smooth connection between your development environment and the W&B platform.

Start by initializing the Weights & Biases environment in your workspace. You can do this by running the following command and following the prompted instructions.

Initial SDK Setup
import wandb

# Initialize your Weights & Biases environment (prompts for the API key, or reads WANDB_API_KEY)
wandb.login()

Navigate to the Weights & Biases authorization page to create and retrieve your API key. Use this key when prompted to authenticate your environment with W&B.

Usage: Training YOLO26 with Weights & Biases#

Usage: Training YOLO26 with Weights & Biases
from ultralytics import YOLO

# Load a YOLO model
model = YOLO("yolo26n.pt")

# Train and Fine-Tune the Model
model.train(data="coco8.yaml", epochs=5, project="ultralytics", name="yolo26n")

Runs that Ultralytics creates chart training metrics against epoch. If you start a run with wandb.init() before training, Ultralytics logs into that run, with an epoch value in every row, and leaves it open so you can keep logging to it and call wandb.finish() yourself. Training arguments are added to that run's config, and any key you or a W&B sweep already set keeps its value.

W&B Arguments#

ArgumentDefaultDescription
projectNoneSpecifies the name of the project logged locally and in W&B, grouping multiple runs together. If unset, W&B runs are logged to the Ultralytics project.
nameNoneThe name of the training run. This determines the name used to create subfolders and the name used for W&B logging.
saveTrueDetermines whether to upload the best checkpoint as a W&B model artifact.
Enable or Disable Weights & Biases

If you want to enable or disable Weights & Biases logging in Ultralytics, you can use the yolo settings command. By default, Weights & Biases logging is disabled.

# Enable Weights & Biases logging
yolo settings wandb=True

# Disable Weights & Biases logging
yolo settings wandb=False
Offline Mode and Resume

Set WANDB_MODE=offline to log without a network connection, then upload the run with wandb sync <save_dir>/wandb/offline-run-*. Resuming training with resume=True continues the same W&B run, online or offline and on every operating system, by reading the run ID from the most recent run folder in the original save directory's wandb/ directory, so keep that directory in place.

Understanding the Output#

Upon running the usage code snippet above, you can expect the following key outputs:

  • The setup of a new run with its unique ID, indicating the start of the training process.
  • A concise summary of the model's structure, including the number of layers and parameters.
  • Regular updates on important metrics such as the training losses, precision, recall, and mAP scores during each training epoch.
  • At the end of training, detailed metrics including the model's inference speed, and overall accuracy metrics are displayed.
  • Links to the Weights & Biases dashboard for in-depth analysis and visualization of the training process, along with information on local log file locations.

Viewing the Weights & Biases Dashboard#

After running the usage code snippet, you can access the Weights & Biases (W&B) dashboard through the provided link in the output. This dashboard offers a comprehensive view of your model's training process with YOLO26.

Key Features of the Weights & Biases Dashboard#

  • Real-Time Metrics Tracking: Observe metrics like loss, accuracy, and validation scores as they evolve during the training, offering immediate insights for model tuning. See how experiments are tracked using Weights & Biases.

  • Hyperparameter Comparison: Every run logs its training arguments, such as learning rate and batch size, so you can compare configurations across runs. For automated search with W&B logging, use Ray Tune.

  • Comparative Analysis: The platform allows side-by-side comparisons of different training runs, essential for assessing the impact of various model configurations and understanding which changes improve performance.

  • Visualization of Training Progress: Graphical representations of key metrics provide an intuitive understanding of the model's performance across epochs. See how Weights & Biases helps you visualize validation results.

  • Resource Monitoring: Keep track of CPU, GPU, and memory usage to optimize the efficiency of the training process and identify potential bottlenecks in your workflow.

  • Model Artifacts Management: Access and share the best model checkpoint, uploaded as a W&B artifact at the end of training, facilitating easy deployment and collaboration with team members on complex projects.

By using these features, you can track, analyze, and optimize your YOLO26 model's training for your object detection tasks.

Summary#

This guide covered the Ultralytics YOLO integration with Weights & Biases, which tracks and visualizes model training results. With W&B, you can streamline your machine learning workflow, make data-driven decisions, and improve your model's performance.

For further details, explore Soumik Rakshit's presentation from YOLO VISION 2023 on this integration.

FAQ#

  • To integrate Weights & Biases with Ultralytics YOLO26:

    1. Install the required packages:

      pip install -U ultralytics wandb
      yolo settings wandb=True
    2. Log in to your Weights & Biases account:

      wandb login
    3. Train your YOLO26 model with W&B logging enabled:

      from ultralytics import YOLO
      
      model = YOLO("yolo26n.pt")
      model.train(data="coco8.yaml", epochs=5, project="ultralytics", name="yolo26n")

    This will automatically log metrics, hyperparameters, and model artifacts to your W&B project.

  • The key features include:

    • Real-time metrics tracking during training
    • Hyperparameter comparison across runs
    • Comparative analysis of different training runs
    • Visualization of training progress through graphs
    • Resource monitoring (CPU, GPU, memory usage)
    • Model artifacts management and sharing

    These features help in tracking experiments, optimizing models, and collaborating more effectively on YOLO26 projects.

  • After running your training script with W&B integration:

    1. A link to your W&B dashboard will be provided in the console output.
    2. Click on the link or go to wandb.ai and log in to your account.
    3. Navigate to your project to view detailed metrics, visualizations, and model performance data.

    The dashboard offers insights into your model's training process, allowing you to analyze and improve your YOLO26 models effectively.

  • Yes, you can disable W&B logging using the following command:

    yolo settings wandb=False

    To re-enable logging, use:

    yolo settings wandb=True

    This allows you to control when you want to use W&B logging without modifying your training scripts.

  • Weights & Biases helps optimize YOLO26 models by:

    1. Providing detailed visualizations of training metrics
    2. Enabling easy comparison between different model versions
    3. Logging the training arguments of every run for hyperparameter tuning comparisons
    4. Allowing for collaborative analysis of model performance
    5. Facilitating easy sharing of model artifacts and results

    These features help researchers and developers iterate faster and make data-driven decisions to improve their YOLO26 models.

Comments25

Ask a question, share your results, or help someone else.

  1. TJdeWetGitHub

    Hi there,

    I would like to know how (if possible) to log more than just these standard metrics for a yolov8 or v11 model during HP tuning logged with wandb: train/loss, val/loss, metrics/accuracy_top1, metrics_accuracy_top5.

    This is what the code looks like:

    !pip install -U ultralytics wandb !yolo settings wandb=True import wandb from ultralytics import YOLO wandb.login(key="...")

    sweep_config = { 'method': "bayes", 'metric': { 'name': 'val/loss', 'goal': 'minimize', }, 'parameters': { .... } }

    sweep_id = ...

    def train(): run = wandb.init() model = YOLO('yolo11n-cls.pt')

    results = model.train(
        data='...',
        batch=wandb.config.batch,
        lr0=wandb.config.lr0,
        epochs=...,
        optimizer=...,
        momentum=...,
        patience=...,
        save_period=...,
        imgsz=...,
        cache=False,
        save=True
    )

    wandb.agent(sweep_id, train, count=...)

    1. Glenn JocherUltralytics

      @TJdeWet for custom metric logging beyond default training metrics in YOLO11 with W&B, you can access the complete results dictionary during training and log additional metrics manually through W&B's API. The automatically logged metrics vary by task type (detection/classification), but you can extend them using a callback. For implementation details, refer to the Weights & Biases integration documentation.

  2. marcory-hubGitHub
    Integration of Weights & Biases with YOLO11 in Google Colab

    Is it currently possible to integrate Weights & Biases (W&B) with Ultralytics YOLO11 in Google Colab? If yes, where can I find the correct and up-to-date information to set this up properly?

    My Issue

    With YOLOv10, I was able to track all my training metrics in W&B seamlessly. However, when trying to replicate this with YOLO11, only GPU and system-related data are logged to the W&B platform. None of the training metrics or results appear.

    Summary of things I've Tried
    • Following the YouTube tutorial linked at the top of this page.
    • Using the Try on Colab button on W&B's site
    WARNING: Did not find branch or tag 'feat/ultralytics', assuming revision or ref.
      error: subprocess-exited-with-error
      
      Ɨ git checkout -q feat/ultralytics did not run successfully.
      │ exit code: 1
      ╰─> See above for output.
      
      note: This error originates from a subprocess, and is likely not a problem with pip.
    error: subprocess-exited-with-error
    
    Ɨ git checkout -q feat/ultralytics did not run successfully.
    │ exit code: 1
    ╰─> See above for output.
    
    note: This error originates from a subprocess, and is likely not a problem with pip.
    Note: The Ultralytcs integration will be soon available as a fully supported feature on Weights & Biases once this pull request is merged.

    This is because the branch https://github.com/wandb/wandb@feat/ultralytics does not exist. To resolve this, I tried various fixes, including following older tutorials and experimenting with suggestions from forums and community discussions.

    I’m particularly keen to switch to YOLO11 because it shows significant improvements in wildlife detection when compared to YOLOv10.

    Any guidance, updates, or resources on enabling this integration would be greatly appreciated!

    1. Glenn JocherUltralytics

      @marcory-hub yes, you can integrate Weights & Biases (W&B) with YOLO11 in Google Colab. Please ensure the latest ultralytics and wandb packages are installed using pip install -U ultralytics wandb. If training metrics are not being logged, ensure your W&B environment is authenticated (wandb.login(key="<API_KEY>")) and that project and name parameters are correctly specified in model.train(). Detailed setup instructions are available in our documentation: https://docs.ultralytics.com/integrations/weights-biases/. Let us know if further assistance is needed!

    2. marcory-hubGitHub

      Thanks for the suggestions, @glenn-jocker ! I tried starting from scratch with W&B with YOLO11 but unfortunately, only system information showed up in W&B.

      On top of that, it seems like my Colab runtime might be banned due to a potential issue with a compromised package. Hopefully, they'll restore access soon for everyone affected.

    3. marcory-hubGitHub

      Good news! My initial code now works for YOLO11. All metrics are automatically saved to W&B, and training is significantly faster than last week without maxing out Colab's RAM.

      I'm not sure why it's significantly faster today, but I suspect the previous slowness prevented training from progressing beyond the warmup epochs, resulting in only system information being sent to W&B.

    4. Glenn JocherUltralytics

      @marcory-hub glad to hear it's working well now, and thank you for sharing your experience! The faster training could indeed be related to runtime or hardware optimizations in Colab. Feel free to explore additional tips and resources in the YOLO11 W&B integration guide to further enhance your workflow. Let us know if you run into any other issues!

  3. Pushtogithub23GitHub

    I used this code, but it only logs system logs; no training metrics, media, or validation metrics are being logged as shown in the video. How can I log all the training metrics and media as demonstrated in the video? Can anyone help me with this?

    from ultralytics import YOLO
    from wandb.integration.ultralytics import add_wandb_callback
    
    wandb.login()
    wandb.init(project = 'Hand-Gestures-Detection', name = 'yolov8n_5',  job_type='training')
    
    model = YOLO("yolov8n.pt")
    
    add_wandb_callback(model, enable_model_checkpointing=True)
    model.train(data="HAND-GESTURE-DETECTION-1/data.yaml", 
                epochs=5,
                imgsz = 640,
                project = "Hand-Gestures-Detection",
                name = 'yolov8n_5'
                )
    1. Glenn JocherUltralytics

      @Pushtogithub23 to log all training metrics and media with Weights & Biases, ensure you have the latest versions of ultralytics and wandb installed. Also, verify that your Weights & Biases account is properly authenticated. For detailed setup instructions, visit the Weights & Biases integration guide.

  4. TJdeWetGitHub

    Hi!

    I am using wandb to do hyperparameter tuning. Is it possible to maximise validation mAP50? This seems to be maximising the training mAP50 if I am not mistaken?

    metric: name: mAP50 goal: maximise

    Otherwise should I do hyperparameter using something like this instead:

    metric: name: val/box_loss goal: minimize

    Thanks!

    1. Glenn JocherUltralytics(edited)

      Hi there!

      For your hyperparameter tuning in Weights & Biases, if you're aiming to maximize validation mAP50 specifically, I recommend adjusting your metric settings accordingly. To focus on validation mAP50, your metric configuration might look something like this:

      metric:
        name: metrics/mAP50(B)
        goal: maximize

      This setup ensures you're targeting the maximization of mAP50 on the validation data directly.

      Using val/box_loss with the goal to minimize can also be effective, especially if you prefer to focus on reducing the box prediction errors during validation:

      metric:
        name: val/box_loss
        goal: minimize

      Both approaches are valid and can be chosen based on your specific model performance goals. Hope this helps! If you have more questions, feel free to ask. 😊

    2. TJdeWetGitHub

      Hi Glenn,

      Thanks for getting back to me. I just tried it, and it doesn't seem that tuning to maximize val/mAP50 is working - the only metrics I see that are being logged are: cls, box, and dfl loss (both train and val), and metrics/mAP50(B), metrics/mAP50-75(B), metrics/Recall(B), and metrics/Precision(B), which I believe are the training metrics? I would like to log those same metrics for validation, and tune based on val/mAP50. Is there a way I can do this?

      Thanks!

    3. Glenn JocherUltralytics(edited)

      Hi there!

      In the Ultralytics W&B integration, validation metrics are logged automatically under names such as metrics/mAP50(B) and metrics/mAP50-95(B) (not val/mAP50; the val/ prefix is used for validation losses like val/box_loss). Use that name in your sweep config:

      metric:
        name: metrics/mAP50(B)
        goal: maximize

      Enable the integration with yolo settings wandb=True; no manual wandb.log() call is needed.

  5. JiyongBooGitHub

    Is it possible to train RT-DETR model with wandb? In add_wandb_callback() function, there are nothing about rt-detr. I tried replacing yolov8 with rt-detr in the example code. In the result, I confirmed that runs.summary["Validation-Table"] and runs.summary["Train-Table"] did not appear. How do I solve this problem?

    1. Paula DerrengerUltralytics(edited)

      @JiyongBoo hi there! 😊 RT-DETR trains with the same built-in Weights & Biases integration as other Ultralytics models, so you don't need add_wandb_callback() or manual wandb.log() calls. Enable it with yolo settings wandb=True, then train as usual:

      from ultralytics import RTDETR
      
      model = RTDETR("rtdetr-l.pt")
      model.train(data="coco8.yaml", epochs=5, project="ultralytics", name="rtdetr-l")

      See the Weights & Biases integration guide for details.

  6. DungHD-1997GitHub

    from ultralytics import YOLO from wandb.integration.ultralytics import add_wandb_callback import wandb def train(): wandb.login() project_name = "test_ai" # Step 1: Initialize a Weights & Biases run wandb.init(project=project_name, job_type="training",name="old_data")

    # Step 2: Define the YOLOv8 Model and Dataset
    model_name = "yolov8x"
    dataset_name = "C://Users//Administrator//project//yolov5_train_data//data//slant_2024.yaml"
    model = YOLO(f"{model_name}.pt")
    
    # Step 3: Add W&B Callback for Ultralytics
    add_wandb_callback(model, enable_model_checkpointing=True)
    model_name = "train_01"
    # Step 4: Train and Fine-Tune the Model
    model.train(project=project_name,name=model_name, data=dataset_name, epochs=100, imgsz=640, batch = 8)
    
    # Step 5: Validate the Model
    model.val()
    
    # Step 6: Perform Inference and Log Results
    #model(["path/to/image1", "path/to/image2"])
    
    # Step 7: Finalize the W&B Run
    wandb.finish()

    if name == "main": train() I used the above code to train custom data, but when using wandb, the GPU memory continuously increases over epochs until an out of memory error appears. program has been stopped. Error message: torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 20.00 MiB. GPU 0 has a total capacity of 15.88 GiB of which 7.62 MiB is free. Of the allocated memory 14.53 GiB is allocated by PyTorch, and 103.44 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)

    1. DungHD-1997GitHub

      If I don't use wandb, the GPU memory hardly changes over each epoch. Is there something wrong with my code?

    2. Glenn JocherUltralytics(edited)

      Hey there! It looks like you're encountering a memory issue related to using Weights & Biases (W&B) with YOLOv8, specifically the GPU running out of memory during training. This can happen for a few reasons, including excessive memory allocation by PyTorch or W&B logging mechanisms. Here are a couple of suggestions that might help:

      1. Reduce Batch Size: Lowering the batch size can significantly reduce memory usage. Although this might increase training time, it can help avoid out-of-memory errors.

        model.train(batch=4)  # Try reducing the batch size
      2. Clear CUDA Cache: Manually clearing the CUDA cache after each epoch might help release unused memory.

        import torch
        torch.cuda.empty_cache()
      3. Optimize Model and DataLoader: Ensure that your DataLoader is not the source of the memory leak, and try to optimize your model's memory usage.

      4. Disable W&B to isolate the cause: Run yolo settings wandb=False and train again; if the out-of-memory error persists, W&B is not the cause.

      Finally, consider profiling your code to identify specific lines or operations that cause the memory spike. PyTorch and W&B offer tools and documentation on best practices to optimize memory usage. šŸ“˜

      Hope this helps! If the issue persists, feel free to reach out with more details or check the W&B and YOLOv8 docs for more insights.

  7. danielst4GitHub

    Hey there, I am trying to make a sweep using WandB but cant quite put together where to insert what into the YOLO call, to use different hyperparameters and track them. Even though I found a site where one has done this he too doesnt describe what he had to do. I am not sure if it is more complicated than I think and if I really have to insert stuff into the training function or if there is an easy way to achieve this. I would be really greatful for help to know what exactly is necessary to make a WandB hyperparameter sweep with YOLO Thanks Daniel

    1. Paula DerrengerUltralytics(edited)

      Hey Daniel šŸ‘‹,

      Great to hear you're exploring hyperparameter sweeps with WandB and YOLO! It's simpler than it might seem at first. Here's a quick guide to get you started:

      1. Initialize WandB: Before starting your training, make sure to log in to WandB in your script or notebook using:
        import wandb
        wandb.login()
      2. Set Up the Sweep: Define your sweep configuration in WandB, using Ultralytics training argument names for the hyperparameters. For example:
        method: random
        metric:
          name: val/box_loss
          goal: minimize
        parameters:
          lr0:
            min: 0.0001
            max: 0.001
          batch:
            values: [16, 32, 64]
      3. Integrate Sweep with YOLO: Enable the built-in integration with yolo settings wandb=True. If a W&B run is already active (as in a sweep agent), Ultralytics logs into it and keeps any config values the sweep already set.
        import wandb
        from ultralytics import YOLO
        
        
        def train():
            wandb.init(project="your_project_name")
            model = YOLO("yolov8n.pt")
            model.train(data="coco128.yaml", epochs=5, imgsz=640, lr0=wandb.config.lr0, batch=wandb.config.batch)
      4. Start the Sweep: Once everything is set up, you can start the sweep in WandB. This will iterate over the defined hyperparameter space, automatically logging each experiment's results.

      Remember, the key parts are enabling the W&B setting and making sure your sweep parameter names match Ultralytics training arguments.

      Hope this helps! If you have any more questions or run into issues, feel free to ask. Happy sweeping! šŸš€

    2. danielst4GitHub

      Thank you Paula, this helped a lot! I can make a sweep now but I noticed that the values of the parameters which are given from the sweep to the script dont match the parameters used. Even though the command looks like this: wandb.wandb_agent - INFO - About to run command: /usr/bin/env python training_rtdetr.py --batch=8 --epochs=6 --imgsz=1280 --learning_rate=0.0007013348811086657 where "training_rtdetr.py" is a script with the code you gave me. (I am working with the RTDETR but it shouldnt make a difference as the metrics and calls are the same if I understand correctly?) But then after engine/trainer it says its using epochs=100 and it also uses that number for training. So just to be sure about a few things, I already logged in to WandB, I wrote a script with the code from your 3., I went to wandb, created a sweep there with the code from 2. + the line program: training_rtdetr.py and in a terminal I used the resulting call wandb agent [sweepid] Also I am kinda confused because you used "learning_rate" and "batch_size" but as far as I can tell these are not the parameters I can give for the training and they should be called "lr0" and "batch" respectively (which I both tried but they didnt work) I hope I provided all the necessary information otherwise just tell me what you need to know

    3. Paula DerrengerUltralytics

      @danielst4 hey Daniel,

      It sounds like you're on the right track with setting up your WandB sweep with YOLOv8! Regarding the mismatch in parameter values, it seems like there might be a discrepancy between the parameter names expected by your script and those defined in your WandB sweep configuration.

      For YOLOv8, the correct parameter names for learning rate and batch size are indeed lr0 and batch, respectively. If you're using these parameters in your training_rtdetr.py script, ensure they're correctly parsed from the command line arguments. Here's a quick Python snippet using argparse for reference:

      import argparse
      
      # Parse command line arguments
      parser = argparse.ArgumentParser()
      parser.add_argument('--batch', type=int, help='batch size')
      parser.add_argument('--epochs', type=int, help='number of epochs')
      parser.add_argument('--imgsz', type=int, help='image size')
      parser.add_argument('--learning_rate', type=float, help='learning rate')
      args = parser.parse_args()
      
      # Now you can use args.batch, args.epochs, etc., in your script

      Make sure your WandB sweep configuration uses the same parameter names as those your script expects. For example:

      program: training_rtdetr.py
      method: bayes
      metric:
        name: val/box_loss
        goal: minimize
      parameters:
        learning_rate:
          min: 0.0001
          max: 0.001
        batch:
          values: [8, 16, 32]
        epochs:
          value: 6
        imgsz:
          value: 1280

      If you've already done this and the issue persists, it might be helpful to double-check how your script handles these parameters and ensure they're being correctly applied to your YOLOv8 model configuration.

      Feel free to share more details if you're still encountering issues. We're here to help! šŸš€

  8. J0ELhGitHub

    Is it possible to remove logging and only log progress in the command line? For my training process yolov8 seems to automatically connect with my wandb account and start logging. This process is very time-consuming so I'd like to know whether there's a flag like log = False or something like that.

    1. J0ELhGitHub

      ^ this is because I'm done experimenting and need quick retraining options now.

    2. Glenn JocherUltralytics(edited)

      Hey there! šŸ‘‹ If you're looking to disable automatic logging to Weights & Biases (W&B) during your YOLOv8 training to speed things up, you can do so by setting the wandb argument to False when initiating your training command. Here's how you can do it:

      yolo settings wandb=False

      This is a global Ultralytics setting, not a train argument, and it stops YOLOv8 from connecting to your W&B account and logging training progress there. Happy training! šŸš€